Reviewed by Ksenia Shilikhina, Voronezh State University
Quantitative methods in cognitive semantics explores the potential and limits of corpus data and statistical methods in semantics. In the two introductory chapters, Dylan Glynn presents an overview of corpus-driven cognitive semantics and Kerstin Fischer reviews quantitative methods in linguistics.
Section 1, ‘Corpus methods in cognitive semantics’, opens with Dirk Geeraerts’s article, ‘The doctor and the semantician’, a methodological discussion of empirical versus introspection-based research traditions in semantics. Though the need for empirical research in cognitive semantics is evident, John Newman’s article, ‘Balancing acts: Empirical pursuits in cognitive linguistics’, claims that usage-based practices and methodologies in cognitive linguistics have their limits and should not be overestimated. In ‘Does frequency in text instantiate entrenchment in the cognitive system?’, Hans-Jörg Schmid shows that the idea of frequency and its relation to the cognitive importance of linguistic elements is not fully understood.
Section 2, ‘Advancing the science: Theoretical questions’, addresses the issues of statistical analysis of grammatical patterns. Stefan Fuhs applies collostructional analysis to English durative constructions in his article, ‘The aspectual coercion of the English durative adverbial’, and Martin Hilpert uses the same statistical procedure in ‘The force dynamics of English complement clauses: A collostructional analysis’.
In ‘Accounting for the role of situation in language use in a cognitive semantic representation of sentence mood’, Kerstin Fischer applies quantitative methods and embodied construction grammar to see how situation influences language use in human-robot interaction. Arne Zeschel’s ‘Exemplars and analogy: Semantic extension in constructional networks’ is an examination of collocations of the German adjective tief with nouns.
Section 3, ‘Advancing the science: Methodological questions’, opens with Stefanie Wulff’s contribution, ‘Marrying cognitive-linguistic theory and corpus-based methods: On the compositionality of English V NP-idioms’, which proposes a special compositionality measure for the semantic contributions of the components of these idioms. Dylan Glynn shows how statistical analysis helps in operationalization and verification of corpus data in ‘Testing the hypothesis: Objectivity and verification in usage-based Cognitive Semantics’. Timothy Colleman examines the semantic range of verbs attracted by the aan-Dative construction in ‘Beyond the dative alternation: The semantics of the Dutch aan-dative’. Dagmar Divjak uses corpus data and statistical analysis to offer a new perspective on Russian imperfective and perfective infinitives in her article, ‘Corpus-based evidence for an idiosyncratic aspect-modality relation in Russian’.
Section 4, ‘Towards an empirical cognitive semantics’, examines the general role of quantitative analysis in usage-based language research. Stefan Th. Gries and Dagmar Divjak address critical comments and objections against quantitative corpus research in their contribution, ‘Quantitative approaches in usage-based cognitive semantics: Myths, erroneous assumptions, and a proposal’. The final article of the volume, ‘Empirical cognitive semantics: Some thoughts’ by Anatol Stefanowitsch, looks at the future of cognitive semantics and discusses the problem of accepting empirical methods into the field.
The volume addresses fundamental questions of empirical approaches to language, e.g. to what degree does corpus reflect human linguistic experience, or how are statistical data to be interpreted in linguistic terms? This book offers a good overview of the advantages and shortcomings of a corpus approach and statistical analysis in cognitive semantics.